We have previously shown that human monocyte-derived macrophages (MDM) infected with HIV-1 and maintained in the absence of exogenous macrophage colony stimulating factor (M-CSF) produce endogenous M-CSF at levels substantially higher than uninfected cells. The kinetics and level of M-CSF production paralleled both the rate and amount of HIV-1 produced. In contrast, production of the cytokines TNF, IL-1, IL-6 and GM- CSF was not observed during in vitro infection of the MDM. Enhanced production of M-CSF was dependent on active replication of HIV-1 within the culture system. We have recently found that treatment of HIV-infected MDM with the drug, AZT, not only leads to complete inhibition of HIV-1 RT activity, but also ablates the production of M-CSF, further supporting a direct role for HIV in enhanced M-CSF production. This suggests that endogenous production of M-CSF can contribute to the survival of HIV-infected monocyte-derived macrophages, enable them to function as a reservoir for the virus, and facilitate the spread of virus in vivo. Work is ongoing to determine potential cellular and molecular mechanisms responsible for this phenomenon, with emphasis on regulation by specific HIV proteins. We are also investigating the correlation between the effects of select cytokines, such as IL-2, TGF-beta, IL-4 and IL-10, on M-CSF production and the effects observed on HIV-1 replication in human MDM. Our results could ultimately lead to new therapeutic approaches directed at eliminating or controlling the expression of HIV in human monocytes/macrophages (MO), thereby reducing the possibility of transmission of virus to the more susceptible CD4+ T lymphocytes. Inasmuch as AIDS is characterized by a general dysregulation of cytokine production, and we have implicated the involvement of endothelin-1 (ET-1) in this disease, we are also investigating the regulation of ET-1 gene expression by various cytokines involved in immune responses. We have found that the cytokine, interferon-gamma (IFN-g) is capable of inducing expression of ET-1 in human MO in a concentration-dependent manner, but expression occurs late compared to that observed in response to the inducer, PMA, indicating that induction of another cellular protein may precede the induction of ET-1 by IFN-g. Although our ability to inhibit the induction of ET-1 by IFN-g by soluble TNF receptors suggested a role for TNF, this cytokine did not directly induce ET-1 production, but enhanced ET-1 production caused by IFN-g. In addition, increased TNF gene expression by IFN-g was not observed. We are currently examining the roles TNF receptors may play in this response. Thus far, our studies suggest that TNF receptors are key mediators of this process and raise the possibility of the therapeutic use of soluble TNF receptors in ET-1-mediated pathologic conditions.